The long term objective is to clarify at the chemical level the form, fate, and function of selenium in the diet, in its role as an essential nutrient and as a modifier of environmental hazards. There is great interest in supplementing the human dietary intake of selenium to correct nutritional deficiencies, or in regard to intervention with the objective of modifying the carcinogenic process. The most critical medical use for selenium is to correct the deficiency that develops with long term total parenteral nutrition. Because of the limited knowledge about forms of selenium, metabolic transformations, and detoxification or excretory metabolites, the present study aims to identify forms of selenium that are excreted in human urine, and clarify the quantitative importance and metabolic origins of trimethylselenonium, the only known urinary metabolite. Another objective is to investigate the pathways by which known forms of selenium that are consumed by humans in food (selenoamino acids) or supplemental forms (sodium selenite) are metabolized to excretory products. The basis for the instability of sodium selenite added to TPN formulas will be investigated so that a better basis for correcting Se deficiency in long term uses of TPN can be developed. A cadmium-binding protein that may be involved in cadmium-induced injury to the microvascular system of testis, which is prevented by selenium, will be purified further and characterized. The methods for fractionating urine involve initial separation into cationic and anionic forms. Separation of cationic compounds will be done by HPLC. An important strategy is to reduce all urinary inorganic Se and compounds of the type R-Se(O)nH(n=0-2) to selenols, then convert these to R-Se-DNP derivatives for characterization by HPLC and mass spectrometry. Once the nature of R is established, model compounds of the type R-Se(0)nH will be synthesized for comparison to the unknown urinary metabolites.